Volume 14 | Issue 6 |

 Abstract

AI-generated synthetic media is becoming a serious threat to digital trust, identity verification, and information integrity across social, legal, and security-critical contexts. Most existing AI detectors analyze only one modality at a time, which leaves them vulnerable when image and audio manipulations are coordinated across channels. This paper presents DeepShield, a multimodal AI detector for image and audio that processes image and audio inputs through separate deep learning pipelines and combines their outputs through late fusion. The visual branch integrates an Xception-based convolutional backbone [1] with a Vision Transformer (ViT-B/16) [6] and frequency-domain feature extraction to highlight spatial and spectral artifacts.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

AI detection for image and audio, multimodal learning, late fusion, Xception, Vision Transformer, MFCC, CNN-LSTM, explainable AI

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

ABSTRACT This is an article on contemporary Gujarati short stories, its transformation, ideologies, subjects, themes, women writers and postmodernism. Particularly, the stories written in 90s and new millennium.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Keyword: Short stories, Gujarati stories, modernism, post-modernism, feminism, post colonialism

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The study examines the influence of the Union Budget on stock market behavior across selected sectors and highlights its importance in understanding how fiscal policy announcements shape investment decisions, stock price movements, and market confidence. The main purpose of this study is to analyse the extent to which budget announcements affect stock prices, returns, and sectoral market performance during the budget period. The objectives of the study aim to identify changes in stock prices and returns of selected companies during the budget period and to analyse the effects of the Union Budget on sector-wise market performance in the pre- and post-budget periods. The research adopts a descriptive research design and is based on secondary data collected from reliable and authentic sources to provide a systematic understanding of market reactions, while purposive sampling is used to select relevant companies and sectors based on the objectives of the study. Data analysis is carried out using ANOVA to evaluate variations and significant differences in stock performance, supported through the use of JAMOVI software for statistical analysis. The findings of this study are expected to contribute to investors, researchers, financial analysts, policymakers, and market participants by offering insights into budget-driven sectoral trends and improving investment decision-making and policy understanding.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Union Budget, Stock Market Performance, Stock Prices, Stock Returns, Sectoral Analysis, Fiscal Policy, Investment Decisions, Market Behavior, Pre- and Post-Budget Analysis.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Corporate Social Responsibility (CSR) has emerged as an essential element of corporate governance, promoting sustainable growth by integrating commercial strategy with social and environmental obligations. The Companies Act, 2013 in India requires companies that satisfy certain financial criteria to donate a minimum of 2% of their average net earnings to CSR activities, representing a notable change in corporate responsibility. This study analyzes CSR spending trends in Bengaluru Urban from 2014 to 2022, emphasizing sectoral distributions like Education, Health, Environment, and Rural Development. The research employs trend analysis, sector-specific distribution, Compound Annual Growth Rate (CAGR), correlation analysis, and comparisons between pre and post-pandemic periods to identify significant changes in CSR priorities. The results demonstrate that CSR expenditures on Education, Health, and Environmental Conservation had a significant rise after 2020, indicating business response to socio-economic concerns. Furthermore, correlation research underscores the strong relationship between CSR expenditure and significant social impact metrics, so affirming its contribution to regional development. This study offers empirical insights for policymakers, enterprises, and scholars, highlighting the need for deliberate CSR investments to optimize long-term sustainability and social effect.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Corporate Social Responsibility (CSR), Bengaluru Urban, Post-Pandemic, Health Infrastructure, Education, Digital Inclusion.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The rapid growth of digital education platforms has transformed the way students access learning resources; however, most traditional e-learning systems still follow a generalized teaching approach that does not adapt to the individual needs, abilities, and learning pace of students. To address this limitation, this paper presents LearnSphere, an AI-powered adaptive learning ecosystem designed to deliver a personalized and intelligent educational experience. The proposed system integrates artificial intelligence techniques such as student classification, recommendation systems, Natural Language Processing (NLP), and Retrieval-Augmented Generation (RAG) to create a dynamic and student-centric learning environment. LearnSphere evaluates student performance through assessments and categorizes learners into different proficiency levels such as Beginner, Intermediate, and Advanced. Based on the learner's performance, preferences, and interaction history, the system recommends customized study materials including notes, videos, and practice questions.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Adaptive Learning, Artificial Intelligence in Education, Personalized Learning, Retrieval-Augmented Generation (RAG), Natural Language Processing (NLP), Recommendation Systems, Learning Analytics, Intelligent Tutoring System.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The Internet of Things (IoT) has become a fundamental component of modern digital infrastructures, enabling interconnected devices to exchange information and support intelligent services across diverse sectors, including healthcare, smart cities, industrial automation, and transportation. While the widespread adoption of IoT technologies has created significant opportunities for innovation and efficiency, it has also introduced serious security concerns. The heterogeneous nature of IoT devices, limited computational resources, and continuous connectivity make these environments attractive targets for counterattacks such as Distributed Denial-of-Service (DDoS), botnet activities, spoofing, malware infections, and data tampering. To address these threats, researchers have increasingly employed Machine Learning (ML) and Deep Learning (DL) techniques within Intrusion Detection Systems (IDS). Although these approaches have demonstrated promising detection capabilities, many of the developed models operate as complex black-box systems, making it difficult for security analysts to understand the reasoning behind their predictions. The lack of transparency can reduce trust in automated security mechanisms and limit their adoption in critical applications where accountability and interchangeability are essential. Explainable Artificial Intelligence (XAI) has emerged as a valuable approach for enhancing the transparency and reliability of AI-based cybersquatting solutions. By providing understandable explanations for model decisions, XAI techniques enable security professionals to identify the factors influencing attack detection outcomes and validate system behavior. This review paper examines recent advancements in dynamic attack detection for IoT environments and explores the integration of explainable methods such as Shapely Additive explanation s (SHAP) and Local Interpreter Model-Agnostic Explanations (LIME) into intrusion detection frameworks. The paper further reviews widely used datasets, machine learning and deep learning models, performance evaluation metrics, and key research challenges. Finally, it highlights emerging trends and future research opportunities aimed at developing adaptive, accurate, and trustworthy IoT security systems capable of addressing evolving caber threats

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Internet of Things (IoT), Intrusion Detection System (IDS), Cybersquatting, Explainable Artificial Intelligence (XAI), Machine Learning, Deep Learning, SHAP, LIME, Dynamic Attack Detection

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The purpose of the present study was to compare sports achievement motivation and aggression among university-level volleyball players according to their playing positions. Volleyball is a highly competitive team sport in which psychological characteristics significantly influence athletic performance and competitive behaviour. Different playing positions in volleyball involve different technical and tactical responsibilities, which may affect the psychological characteristics of players. A total of 105 male university-level volleyball players from different universities of Chhattisgarh were selected randomly as subjects. The selected players were equally divided into three groups consisting of libero, attacker, and setter players. Sports Achievement Motivation Test (SAMT) developed by M. L. Kamlesh (1990) and Sports Aggression Inventory (SAI) developed by Anand Kumar and Prem Shankar Shukla (1998) were used for the collection of data. Mean, standard deviation, Analysis of Variance (ANOVA), and Post Hoc analysis were employed for statistical analysis. The findings of the study revealed statistically significant differences among libero, attacker, and setter volleyball players with respect to achievement motivation and aggression. Attacker players demonstrated the highest achievement motivation and aggression levels, followed by setter players, whereas libero players recorded the lowest scores among the three groups. The study concluded that psychological variables differ according to the playing positions of volleyball players and highlighted the importance of position-specific psychological preparation and mental training in volleyball coaching programmes.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Achievement Motivation, Aggression, Volleyball, Playing Positions, Sports Psychology, University Players

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Underground coal mining involves serious safety risks due to hazardous gases, high temperatures, fires, and structural failures. Delays in detecting these hazards often lead to accidents such as methane explosions and toxic gas exposure. This paper proposes an IoT-enabled real-time safety monitoring and alert system to address these challenges effectively. The system uses an Atmega328 for local processing and a NodeMCU ESP8266 for cloud connectivity, monitoring parameters like temperature, humidity, gas levels, flame, and tilt. It provides instant alerts through buzzers and LEDs while transmitting data to an MQTTbased cloud platform. The proposed system improves early hazard detection, response time, and overall mine safety.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Coal Mine Safety, IoT, Atmega328, NodeMCU, MQTT, Real-time Monitoring, Wireless Sensor Network, Hazard Detection, Mobile Application.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Neem Ka Thana Nagar me Nagariy Prasar: Ek Bhogolik evm Upgrahiya Chitra Aadharit Vishleshn (1995-2025)

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Neem Ka Thana Nagar me Nagariy Prasar: Ek Bhogolik evm Upgrahiya Chitra Aadharit Vishleshn (1995-2025)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Methylxanthines, particularly 1,3,7-trimethylxanthine (caffeine) and its primary in vivo metabolites (theobromine, theophylline, and paraxanthine), represent some of the most ubiquitously consumed psychoactive substances worldwide. Beyond their historical classification as dietary stimulants, modern molecular pharmacology has uncovered an intricate landscape of bioactivities that extend deep into therapeutic and neuroprotective territories. This extensive review outlines the comprehensive pharmacological mechanisms of methylxanthines, focusing heavily on their multi-targeted interactions within the central nervous system (CNS). We dissect the competitive antagonism of adenosine A1 and A2A receptors, the downstream modulation of neurotransmitter cascades (including dopamine, acetylcholine, serotonin, and gamma-aminobutyric acid), and the inhibition of cyclic nucleotide phosphodiesterases (PDEs). Furtherm examine the potent intracellular calcium mobilization and the epigenetic modifications orchestrated via histone deacetylase activation. In the context of neurodegenerative pathology, we present exhaustive evidence supporting caffeine's therapeutic efficacy against Alzheimer's disease, Parkinson's disease, and Amyotrophic Lateral Sclerosis, highlighting its capacity to mitigate neuroinflammation, suppress microglial overactivation, neutralize reactive oxygen species (ROS), and prevent mitochondrial collapse. Despite these profound medicinal attributes, the clinical translation of methylxanthines remains severely bottlenecked by complex pharmacokinetic challenges, including rapid systemic clearance, low metabolic bioavailability, and the restrictive stringency of the blood-brain barrier (BBB). To address these limitations, we comprehensively evaluate cutting-edge nanotechnological intervention strategies. This includes a critical analysis of lipid-based nanoparticles (SLNs and NLCs), polymeric nanoparticles, nanoemulsions, and functionalized carbon nanotubes designed for targeted, sustained, and brain-specific delivery. Ultimately, this review bridges the gap between classic xanthine chemistry and advanced clinical nanomedicine, providing a definitive framework for utilizing methylxanthines as novel, high-efficacy neurotherapeutics.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Methylxanthines, caffeine, Alzheimer's disease, neurotherapeutics, Neuroprotection, and Emerging Nanotechnological Delivery Systems

  License

Creative Commons Attribution 4.0 and The Open Definition